Video streaming is critical in IoT systems, enabling a variety of applications such as traffic monitoring and health caring. Traditional adaptive bitrate streaming (ABR) algorithms mainly focus on improving Internet video streaming quality where network conditions are relatively stable. These approaches, however, suffer from performance degradation at IoT edge. In IoT systems, the wireless channels are prone to interference and malicious attacks, which significantly impacts Quality-of-Experience (QoE) for video streaming applications. In this paper, we propose a secure and network-state-aware solution, SASA, to address these challenges. We first study the buffer-level constraint when increasing bitrate. We then analyze the impact of throughput overestimation in bitrate decisions. Based on these results, SASA is designed to consist of both an offline and an online phase. In the offline phase, SASA precomputes the best configurations of ABR algorithms under various network conditions. In the online phase, SASA adopts an online Bayesian changepoint detection method to detect network changes and apply precomputed configurations to make bitrate decisions. We implement SASA and evaluate its performance using 429 real network traces. We show that the SASA outperforms state-of-the-art ABR algorithms such as RobustMPC and Oboe in the IoT environment through extensive experiments.

中文翻译：

---

在物联网边缘实现安全和网络状态感知比特率自适应

视频流在物联网系统中至关重要，可实现流量监控和健康护理等各种应用。传统的自适应比特率流（ABR）算法主要集中在网络条件相对稳定的情况下提高Internet视频流质量。但是，这些方法在物联网边缘的性能下降。在物联网系统中，无线通道容易受到干扰和恶意攻击，这会严重影响视频流应用程序的体验质量（QoE）。在本文中，我们提出了一种安全且具有网络状态意识的解决方案SASA，以应对这些挑战。我们首先研究增加比特率时的缓冲区级别约束。然后，我们分析吞吐量过高估计对比特率决策的影响。根据这些结果，SASA设计为包含脱机阶段和联机阶段。在离线阶段，SASA会在各种网络条件下预先计算ABR算法的最佳配置。在在线阶段，SASA采用在线贝叶斯变化点检测方法来检测网络变化并应用预先计算的配置来做出比特率决策。我们实施SASA，并使用429条实际网络跟踪评估其性能。我们通过广泛的实验表明，在物联网环境中，SASA的性能优于最先进的ABR算法，如RobustMPC和Oboe。SASA采用在线贝叶斯变化点检测方法来检测网络变化并应用预先计算的配置来做出比特率决策。我们实施SASA，并使用429条实际网络跟踪评估其性能。我们通过广泛的实验表明，在物联网环境中，SASA的性能优于最先进的ABR算法，如RobustMPC和Oboe。SASA采用在线贝叶斯变化点检测方法来检测网络变化并应用预先计算的配置来做出比特率决策。我们实施SASA，并使用429条实际网络跟踪评估其性能。我们通过广泛的实验表明，在物联网环境中，SASA的性能优于最先进的ABR算法，如RobustMPC和Oboe。